re: co-generation is probably overkill for a backup generator
28 dec 2001
>co-generation (where you make use of the otherwise waste heat) can
>substantially improve efficiency,
let's see... honda's $1,069 68 pound 17.3"x15.7"x18.9" eb3000 generator
makes 2,600 w for 6 hours (15.6 kwh) with 2.7 gallons of gasoline, so the
gas-to-electric efficiency is about 100x15.6x3412/(2.7x130k) = 15.2%.
we might raise that to 90% by using the "waste heat" :-)
>but it's way overkill for a simple backup-generator.
i seem to recall a uk water heating system that hangs out a window like
an ac. why let a "simple backup generator" just collect rust most of the
time when we can use it more often? most cogen units are big, but it's
nice to be able to put one into a car trunk and take it to be repaired.
>building a proper "indoor enclosure" for co-generation might be a bit
>beyond the average homeowner.
the enclosure might be as simple as a 2'x4'x6' tall box made with block
walls laid up dry in the basement, and plastic film over that. the genny
might sit on top of 2 55 gallon water drums. the box wouldn't be airtight.
the genny would draw in room air through a small hole in the box for
combustion, and all the exhaust air would end up outdoors, so the box
has a slight negative pressure compared to the room, and exhaust gases
don't end up in the room. block walls can soundproof well.
>you would need to hire somebody with some engineering skills like
>yourself nick, and i'd guess that would be relatively expensive.
hideously expensive, but maybe not needed. then again, we might make
more than one of these things.
>you could undoubtedly make co-generation a bit easier to accomplish
>with an indoor installation because the otherwise waste heat is closer
>to where you need it.
>but in doing so, you create so many new problems to solve that i
>doubt it would be practical for these folks in this application.
let us examine these problems in excruciating detail.
>for an indoor installation...
>i'd want to run the exhaust pipe directly outside.
i'd run it into a stainless steel gas flex hose and a spiraled soft
copper pipe in a 55 gallon drum full of water to cool it to 212 f
before piping it outside. how many feet of pipe? i don't know. haven't
tried it or done any calcs. look at a long pipe. see where it stops
that honda makes 50k btu/h of heat, not counting the electrical output,
about 300k btu over 6 hours, enough to heat 300k/150 = 2k pounds of
water 150 f, about 4 drumfuls. we might use 2 plastic drums inside the
enclosure and 2 more outside, connected by 8 short heater hoses pushed
into holes in the sides near the tops and bottoms.
say half of the heat leaves via a 50 cfm 500 f exhaust and the other half
leaves via the cooling fins. we might move air over the fins with a 500
cfm fan (warming the air by about 25k/500 = 50 f), with an auto radiator
in the airpath below thermosyphoning warm water through a drum with a
25k/800btu/h-f = 31 f air-water temp diff. we might measure the genny fin
temp in still 110 f air, and make sure it doesn't exceed that temp while
inside the box in warmer moving air.
>you want to be 100% sure that you are getting rid of all of the co
>emissions. a co detector is a fine idea, but it just provides an extra
>margin of safety and not the primary means.
sure. it's a detector, not a remover.
>you have to design the enclosure box so that you can access the generator
>for periodic maintenance.
some sort of removable 2'x2' panel or door.
>for co-generation, you need a heat exchanger and ways to control it.
in still air, the 4 drums have about 100 ft^2 of r2/3 surface with
rc = 2/3/100x2000 btu/f = 13.3 hours. in a 70 f room, they would cool
from 200 to 100 f in -13.3ln((100-70)/(200-70)) = 19.5 hours. we might
only fire up the system on very cold days to offset the house heating
need from some other source, or insulate the genny box and build another
box with temp controlled vents around the 2 outer drums. we might use
an rv pump to pressurize drum water for showers, and keep them filled
with cold water from a float valve.
>co-generation is a great way to improve the efficiency and get the
>maximum benefit from ever ounce of fuel, but it takes a fair amount
>of engineering to accomplish...
agreed, but this ain't rocket science.
>being off-grid and heavily dependent on a generator for everyday use, i'm
>the kind of guy who would benefit from co-generation. even so, i'm inclined
>to locate the generator in a out-building specifically designed for that
>purpose. since i run the generator frequently, i'd like to squeeze every bit
>of efficiency out of a unit that i can.
like, 90% vs 15%?